利用三元组损失进行无监督动作分段（英文版）

英文标题：Leveraging triplet loss for unsupervised action segmentation中文摘要：本文提出了一种全新的无监督学习框架来从单一的输入视频中学习适合于行动分割任务的动作表示，无需任何训练数据，达到了与现有无监督方法相比更高质量的时间边界恢复，并通过应用聚类算法在学习的表示上取得了具有竞争力的表现。英文摘要：In this paper, we propose a novel fully unsupervised framework that learnsaction representations suitable for the action segmentation task from thesingle input video itself, without requiring any training data. Our method is adeep metric learning approach rooted in a shallow network with a triplet lossoperating on sim